An Economic Analysis of Bird Damage in Vineyards of the Marlborough Region by Laurie Boyce, Anton Meister and Sandy Lang Discu

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Research Papers in Economics

An Economic Analysis of Bird Damage in Vineyards of the Marlborough Region

Laurie Boyce, Anton Meister and Sandy Lang

Discussion Paper in Natural Resource and Environmental Economics No. 20 Centre for Applied Economics and Policy Studies Department of Applied and International Economics Massey University, Palmerston North New Zealand

March 1999

Foreword

As the area of vineyards in New Zealand keeps on expanding the costs of bird damage will also increase in importance. While currently the loss of production due to bird damage from a national point of view may be insignificant compared to losses due to other factors such as weather and market fluctuations, the cost of bird damage may be the difference between profit and loss to individual growers.

The Marlborough region was chosen as a case study to analyse the impact of bird damage. This region is the largest viticultural region within New Zealand in terms of area and quantity of grapes. A postal survey was conducted to determine the potential loss due to bird damage, the loss avoided through control and the cost of that control. In addition different control measures were evaluated in terms of cost and effectiveness.

Control measures are costly and it is important that the most cost-effective measures are used. This discussion paper provided factual information on control costs, effectiveness, regional impacts and economic impacts and should be of interest to those involved in the wine industry as well as those involved in research in new technologies for bird control. The discussion paper also raises the issue of the wider social costs associated with some of the control measures.

The discussion paper is the result of research conducted by Ms Laurie Boyce for a Bachelor of Applied Economics (Honours) degree, supervised by Professor Anton D Meister from the Department of Applied and International Economics, Massey University and Dr A (Sandy) Lang from HortResearch. Special thanks go to the Marlborough Grape Growers Association, HortResearch, the Marlborough District Council, the Marlborough Research Centre Trust and Montana Wines (Marlborough) for financial and other support.

Anton D Meister Professor, Resource and Environmental Economics

Table of Contents

Foreward ...... ii Table of Contents...... iii

Chapter 1 Introduction...... 1 2 Economic Theory...... 8 3 Methodology...... 11 4 Analysis of Results ...... 21 5 Social Issues...... 45 6 Key Assumptions and Limitations...... 47 7 Conclusions...... 50

Bibliography ...... 52

Appendices ...... 55

Chapter 1

Introduction

1.1 The Problem and the Objectives

“The actual loss of yield to bird and mammal pests in volume and dollars has not been assessed for many crops in New Zealand as it is difficult and expensive to do so” Porter, Rudge and McLennan, 1994, p8

This research project aims to quantify the costs and benefits associated with bird damage and bird control measures. As identified in the quote above, the actual loss due to bird damage is difficult and expensive to assess, this is because it is time consuming, and particularly for grapes, difficult to identify. It is estimated to be about 20% of grapes within a vineyard (Porter et al, 1994). Trim’s 1982 paper reported 10% damage in the Marlborough/Nelson Westcoast region, over all crops.

In the 1987 season, Dick Porter, then a DSIR ecologist, estimated that Hawkes Bay grape growers lost around $1 million in production from grapes damaged by birds. This estimate included money spent on bird damage prevention. He surveyed 6 vineyards and found some spent as much as $742 per hectare on bird damage prevention, and suggested that much of it was to no avail (Hawkes Bay Notes, 1987). Porter believes that bird damage is the biggest problem the industry needs to address and that there is plenty of room for research into the problem. “While there is some basic research data on understanding the problem, what is needed now is research to find out how the damage can be reduced on grapes at a cost which is worthwhile... It is no good spending a thousand dollars just to reduce a hundred dollars worth of damage in a vineyard, and this has happened on some properties where people are spending far more than the level of damage.” Hawkes Bay Notes, 1987, p283 Marlborough was chosen as a case study because of its rapidly expanding importance as a grape growing area.

The objectives of this research are: 1. To determine the relative effectiveness of different bird control methods. 2. To calculate the financial and economic benefit of bird control in vineyards of the Marlborough region. 3. To describe the social issues associated with bird control. 4. To draw conclusions and make recommendations regarding the current state of bird control in Marlborough.

1.2 Background Bird damage in New Zealand is on a relatively low scale compared to some parts of the world because New Zealand’s birds are resident. The population of our birds only varies as they die in winter and breed in summer, although some New Zealand birds have ‘seasonal excursions’ between foothills and plains. Where bird species are migratory, as they are overseas, damage can be catastrophic due to millions of flocking birds that can destroy whole crops (Porter et al, 1994).

The effects of bird damage can still be devastating in New Zealand, depending on weather conditions, fungal rots and diseases which all vary from year to year. There also appears to be a rather obvious dynamic relationship between bird species prevalent in an area, the crops and the damage done (Trim, 1982). Some birds remove the whole grape while others ‘peck’ at the fruit leaving it open to fungal infection. The seeping juice attracts wasps, but Porter et al (1994) suggest that this can be beneficial as they limit the dripping of juice that would otherwise spoil the rest of the fruit and causes fungal rot. In certain weather conditions and if fungal diseases are rife, this ‘secondary’ damage is potentially devastating. Secondary damage decreases the quality of the must1; hence it depreciates the quality of wine produced.

Birds are particularly difficult to manage as ‘pests’ because, unlike rodents and possums, many people regard birds “...as a delight to the eye and ear” (Porter et al, p7). Birds of the countryside and gardens do not appear as pests, so there is an obligation to manage them in an environmentally friendly and humane way. The damage is often highly seasonal and localised, which also makes management difficult.

In the Canterbury region rooks are considered pests of local importance (Canterbury Regional Council, 1991). In 1991 rook control was undertaken by the Council and funded by rates. Although rooks do not damage grapes (see Table 2), flocks of 100 plus have been known to ruin emergent crops in a single day. Rooks mainly eat and destroy newly sown cereals, ripening lentils, walnuts, acorns, and pumpkin seeds (Canterbury Regional Council, 1991, p61). In the past, farmers undertook their own control measures, but it was evident that the main impact was to shift birds to other properties in the area, or cause the birds to shift and create new rookeries. The Manawatu-Wanganui Regional Council poisons thousands of rooks each year. The rooks cause about $120,000 worth of damage to crops in this region each year (Sargent, 1998). These are examples of where regional councils have seen obvious benefits from controlling birds.

The loss of production due to bird damage could be insignificant compared to losses from other factors like the weather, market variation and poor land management. Unfortunately for birds, they are more visible than bad management and therefore more readily blamed. Bird damage may be the difference between profit and loss to the

1 Winepress

5 individual grower, but to any particular industry, region or country as a whole, it may be relatively unimportant.

Marlborough is the largest viticultural region within New Zealand, in terms of area and quantity of grapes. According to the results of the Lincoln University Economic Impact study2 for each $1 increase in the demand for Marlborough wine grapes, $2.07 worth of economic activity is generated. There has been a continual expansion since the late 1980s, 93%, or 1324 ha since 1990, with respect to grape growing area3. It is expected to continue to be New Zealand’s dominant wine grape region, with 37.3% (2747 ha) of the national producing area in 1998. The forecasted producing area for 2001 is 3323 ha4. Hence, the bird damage problem is not expected to show any sign of decline.

Nationally, there has been a significant move by many growers to higher valued varieties of grapes. Marlborough has followed this trend. This implies that crop losses due to bird damage will increase in value. The figures in Table 1 below reflect this trend. The planted area of Muller Thurgau has declined rapidly in New Zealand, from 1306 hectares in 1990 to 537 hectares in 1998, and is predicted to continue doing so. The projected plantings of Muller Thurgau in 2001 is only 353 hectares. Conversely, the planted areas of Sauvignon Blanc and Chardonnay have dramatically increased.

The new varieties are worth significantly more than Muller Thurgau as can be seen in Table 1 below. They also demand a higher level of quality. High quality is also increasingly demanded from standard varieties as buyers of wine grapes aim to increase the quality of their wine. Clauses stating maximum levels of damage to grapes are now common in contracts between growers and buyers. Although some wine grapes are exported out of the region, a vast majority is used in Marlborough wines. The cost of bird damage to individual growers, the region, and the country will probably increase in the future. Likewise, additional economic activity will be generated by the overall increase in production and value.

2 A Summary of the Results of the 1997 Marlborough Wine Industry Economic Impact Study 3 The Bank of New Zealand Wine & Grape Industry Statistical Annual, 1998, p5 4 The Bank of New Zealand Wine & Grape Industry Statistical Annual, 1998, p19

6 Table 1 Comparison of Planted Area and Price between selected Grape Varieties Planted Hectares in Average Price per Planted Hectares Variety NZ Tonne 1998 in Marlborough 1990 1998 2001* NZ Marlborough 1998 2001* Muller Thurgau 1306 537 353 444 477 105 61 Sauvignon Blanc 427 1678 1923 1286 1368 1150 1382 Chardonnay 689 2006 2800 678 818 Still 1428 1481 Sparkling 1397 1416

*forecasted Adapted from The Bank of New Zealand Wine & Grape Industry Statistical Annual 1998

Accurate scientific assessment of the losses requires careful measurements and proper statistical analyses. This is especially difficult for crops that ripen in sequence, such as different varieties of grapes. Damaged fruit could be counted, but this would require detailed inspection as often the whole fruit is removed without leaving evidence unless the stalks are carefully checked. For a grower this type of assessment is out of the question, and the situation is often confused by natural loss and wind damage. Estimation of the secondary damage would also be difficult. The growers can make an educated ‘guess’ at the overall level of damage. However, as Porter et al (1994) note, the actual measurement is often less than this subjective judgement.

Bull (1983) highlights some of the difficulties in assessment. A nation-wide survey of orchard and berry gardens during the 1966-67 growing season5 revealed important defects in data collection by questionnaire. Only 14% of grower’s responded and subsequent interviews with a sample of the non-respondents showed that they had suffered less bird damage than respondents had. Therefore, the questionnaire tended to exaggerate the overall extent of the damage. He also noted that few growers had time to watch birds closely to observe whether they are feeding on unblemished fruit or windfall fruit that is already damaged. The DSIR admitted these limitations and found that the most damaging bird species were blackbird, song thrush, myna, starling, white- eye and the house sparrow. They also found the fruits most chiefly affected were strawberries, grapes, cherries, pears and apples.

Bird control measures are varied in method and effectiveness. Bird control has become an increasing problem since the withdrawal of a chemical, Methiocarb. It was a highly effective method, leaving only 1% damage (Hawkes Bay Notes, 1987). It was removed because of overseas complaints due to unacceptable levels of chemical residue in some

5By the Ecology Division of DSIR (New Zealand Journal of Agricultural Research, May 1970)

7 New Zealand wine. This highlights the growing awareness of environmental and health concerns within society.

A satisfactory bird control measure must be legal and should avoid offending neighbours through excessive noise and loud explosions (Brough, 1962-65). This poses a problem for growers because many effective control measures are noisy. The Acclimatisation Society suggest that bird scarers that imitate gunshot noises are the best deterrent (Trim, 1982).

Increasing urban encroachment onto rural areas creates a conflict of interest that is increasingly hard to deal with. The question arises as to whether the Council needs to include provisions to limit noise from bird control methods. Mr Gregan, the Chief Executive Officer of the Wine Institute, has stressed the importance of the issue from an industry point of view. He said that it is “...imperative appropriate provision for the industry is made by local regulatory authorities, notably under the Resource Management Act, so expansion of the industry is not curtailed in the years ahead”.6

Noise emission from bird scaring devices is currently a contentious issue in the Marlborough Region (Munden, 1998), particularly as the Marlborough District Council is in the process of preparing an Annual Plan. Recently, it became a political issue during the local body elections, and decisions on its inclusion in the Annual Plan were put on hold.

When considering methods of bird control growers need to consider the species of bird causing the damage. Table 2 contains a list of the problem birds and control measures available. While the table covers the most common methods, there are other methods. The effectiveness of some deterrents depends on the biology of the bird. “Mixed flocks of starlings and house sparrows in urban areas will respond quite differently to deterrent sounds” (Porter et al, 1994, p10). For example, Multi-Frequency Generators (MFG’s) are only effective on waxeyes while starlings are deterred by AV-Alarms. Starlings remove more grapes in a short time than other species because they flock, whereas thrushes and blackbirds take just as much, but inconspicuously. Thrushes and blackbirds are more susceptible to trapping, but do not respond to alarm calls.

A combination of shotgun patrols and scatterguns is effective in deterring birds. Airhorns mounted on four wheel motorbikes and foot patrols can also be effective. Movement in and around the vineyard is important, or the birds become accustomed to the noise very quickly. An American study of bird behaviour (Verrengia, 1998) has shown that birds have episodic, or event based memory. Hence, they may be able to recall where, and when the noise had occurred and become accustomed to it. It seems that growers will employ a variety of bird control measures to achieve the best result.

6 A Summary of the Results of the 1997 Marlborough Wine Industry Economic Impact Study

8 Timing of control measures is also critical. Ripened grapes are highly attractive to birds, so control must begin at this time. “Populations of any animal or bird are in dynamic balance with a variety of factors, all of which have to be taken into account for control to be effective” (Trim, 1982, p5).

Table 2 Identification of Problem Birds and Common Control Measures

Birds Identified as a Problem for Blackbird Grapes (in alphabetical order) Greyfinch House Sparrow Myna Rosella Song Thrush Starling Waxeye (Silvereye) Bird Control Measures Commonly AV-Alarm Used (in alphabetical order) Barriers Big-eyed Balls Enclosures Pop-up Scarecrows Protective Measures (netting) Scareguns Shooting Starling Distress And Alarm Calls Trapping Adapted from Porter et al (1994)

Shelterbelts provide a roosting site for birds, so damage will probably be worse along tree lines. Grapes need little shelter so growers tend to remove or thin non-essential shelterbelts on their properties. The edges are also generally the worst affected as birds prefer to have easy access to food. Surrounding buildings, poultry farms, orchards and other features will also influence the areas damaged. Growers may place netting over the rows along tree lines and edges of their vineyards. Netting is considered to be an effective and humane damage prevention measure, but to cover large vineyards the cost can be prohibitive. Growers are concerned too that the quality of grape may be affected

9 due to increased humidity and shading. Research has shown that netting can exacerbate botrytus problems if it is particularly humid7.

Certainly, the costs of each of these control methods, and the level of bird damage they can prevent, needs to be quantified so that growers can make more informed choices about which methods to use. Each grower must be willing to spend at least as much on bird control as the value of the damage it is preventing, but certainly not more. In marginal terms, any more would be uneconomic, as would any less. The optimal level of bird control for each individual vineyard is where the marginal cost equals the marginal benefit of that control. That is, where the cost of an additional unit of control yields an equal additional unit of benefit from damage prevention.

Growers who participated in the survey (to be discussed later) were asked to give information on the costs of bird control and the level of bird damage that still occurs. They were also asked to estimate the damage caused by birds if they employed no bird control methods. By gathering information on the costs of bird control, and growers’ perception of their damage prevented, it is possible to quantify the costs and benefits. Hence, this research involves the use of grower’s estimates of damage rather than the use of accurate scientific assessment, which is inherently difficult, as discussed earlier. Information on the perceived effectiveness of each bird control method was also collected.

7 Trought, personal communication

Chapter 2

Economic Theory

While a financial analysis may be concerned with financial profit or loss from an individual or industry point of view, a benefit cost analysis may consider resource allocation from a national point of view. This leads to differences in the results of each.

2.1 Financial Analysis The financial analysis is from the point of view of individual vineyards. The financial analysis forms the basis for the economic analysis. In this study, information on the primary costs and benefits of bird control was gathered using the postal survey. This included:

Primary Costs Capital costs of bird control equipment Maintenance and annual operating expenses of bird control equipment Hours of labour dedicated to bird control Hours of motorbike, tractor and other vehicles used for bird control

Primary Benefits The value of crop damage avoided

Data collected in the questionnaire were used to evaluate, for each vineyard, the total amount spent on each bird control method used for the 1998 season. These were then compared to the effectiveness ratings given. The costs for each vineyard can then be summed and total regional expenditure on bird control can be calculated. The total cost figures reflect the actual amount spent on bird control by vineyards in the region in terms of:

Total expenditure and per hectare expenditure on individual methods Expenditure on labour The total cost of the bird damage problem to each vineyard

11 From an individual vineyard’s perspective, the total cost of bird damage includes the cost of bird control and the loss of crop value due to remaining bird damage. This relationship will also be examined.

A secondary benefit of bird control is termed the ‘multiplier effect’. The extra output created (because less damaged by birds) and inputs bought all generate flow-on economic activity within the region. The magnitude of this effect is not examined in this analysis, however it is significant from a regional perspective.

2.2 Benefit Cost Analysis A benefit cost analysis involves the concept of economic profitability rather than financial profit. That is, it deals with the real worth of a project to a country.

The basic benefit cost analysis rule is that “...an expenditure is judged to be potentially worthwhile if its benefits exceed its costs, where benefits and costs are defined to include any welfare gain and loss which occurs because of the expenditure on the project” (Pearce, 1983, p25). Here, the ‘project’ is bird control implemented by individual vineyards in the Marlborough region and the analysis is from the point of view of New Zealand.

To evaluate the costs and benefits of the bird control, values from the financial analysis are modified to meet the benefit cost analysis criteria and assumptions.

In a Benefit Cost analysis, ‘transfer payments’ must be removed to reflect the real economic costs and benefits. Transfer payments do not represent flows of actual goods and services within an economy, i.e. they do not constitute a using-up of real resources (such as labour hours) but are merely a redistribution of money. Such a transfer does not add or subtract from the social welfare of a country because it merely transfers money from one person to another. For example, income tax (a loss) is paid to the government and is redistributed to members of society via income support payments (a gain). Other examples of transfer payments include subsidies, debt servicing and interest. These must be removed so prices reflect the true cost to society.

This analysis is from a national point of view so payments of Goods and Services Tax, Income tax and Accident Compensation Levies are removed from regional expenditure because they are ‘transfer payments’. Hence, the ‘cost’ of bird control is the level of regional expenditure minus the GST, Income tax and ACC levy components.

The ‘benefit’ of a project is the value of any costs that have been avoided because of the project. In other words, it is the difference between the ‘with’ and the ‘without’ situations. For this analysis, it is the value of grape damage that has been avoided due to the implementation of bird control measures.

Further to this, there is an external cost; the social cost of bird control that is created by the use of noisy bird control methods. There is no market price for “peace and quiet”,

12 so the value of this cost cannot be easily estimated and has not been included in this analysis. This does not take away the fact that external costs are real costs to society (i.e. cause a decrease in social welfare) and should therefore be included in a Benefit- Cost analysis. There are techniques available to place dollar values on costs of this nature, this however, was not done in this study.

Sometimes individuals are required to internalise the external costs they produce. For example, under the Resource Management Act 1991, regional and local councils can require the mitigation or avoidance of such effect (e.g. noise pollution or water pollution). The internalisation process imposes a cost on the firm required to do the mitigation and hence in a benefit-cost analysis of a bird control project (with noise mitigation requirements) the external cost is no longer external but internal to the firm and is included in the analysis.

There are a variety of ways in which regional and local authorities attempt to deal with the problem of external costs (e.g. negative environmental effects). Policy options are moral suasion, education, outright bans, standards or economic incentives and disincentives. Currently the issue of noise and appropriate policies is being debated in the Marlborough region and this will be discussed further later on in this report.

It should also be recognised that a cost benefit analysis ignores the distribution of costs and benefits. The analysis is based on utilitarianism and its underlying ethic. More specifically, the analysis gives an equal weighting to each individual in society, thereby assuming that each person values money in the same way. This has implications for policy decisions based on cost benefit analysis.

Chapter 3

Methodology

3.1 Data Collection In order to obtain the necessary information, both personal interviews and a postal survey were conducted. An interview with the Marlborough District Council was also conducted. The personal interviews were used to gather preliminary information on which the postal survey was based. A discussion of the results of the personal interviews is presented below.

3.1.1 Personal Interviews Selection of Participants The participants in the personal interviews were chosen on the basis of willingness to participate and availability during the week I was in Blenheim. The 12 vineyards covered a range of vineyard areas and locations; however, they were all in the Marlborough Region.

Development of the Personal Interview Questionnaire A questionnaire was developed to both guide the personal interview and collect preliminary information for the development of the postal survey (see Appendix I). The interviews also provided some good background information, as participants were encouraged to give advice and opinions on the issues involved.

General information about the vineyard and its proximity to shelter belts and buildings etc was gathered to identify common factors and problem areas on each vineyard. The participants were asked which bird control methods were used so that a comprehensive list of methods used could be made. The costs of each method and the estimated level of bird damage with and without bird control were essential questions for this analysis. It was recognised that the costs and benefits of bird control would need to be expressed in comparative monetary values, so it was necessary to ask participants for a crop value and/or turnover for the vineyard.

The personal interviews helped to identify how the postal survey should be structured, which questions should be asked, and how they should be asked. For instance, it was

14 realised that questions on the cost of bird control methods would need to be very specific so that the data could be aggregated and compared easily.

Information on the frequency and magnitude of bird damage was gathered to determine if the 1998 season would be comparable to other years. Participants were also asked of they thought the problem was becoming worse, and why. This was to identify any wider issues that could be important.

Opinions on the secondary effects of bird damage were needed to determine if the postal questionnaire should differentiate between direct loss of fruit and indirect loss through secondary damage. Opinions on the effect of bird damage on the market price for grapes were used to identify whether bird damage has a price effect.

The information on the effectiveness of each bird control technique was necessary for comparing perceived effectiveness and the amount spent on the method. The participants were asked to give an effectiveness rating and a rating of how often the method is used. These ratings were to be given on a scale of 1 to 5, 1 being ‘poor’ or ‘never’, 5 being ‘excellent’ or ‘all the time’.

The interview with the Marlborough District Council was an informal discussion of the social issues involved with bird scaring. The issues identified are included in the discussion below.

3.1.2 Discussion of Personal Interview Results Methods The majority of the 12 growers interviewed used a combination of netting, gas guns and persons shooting around the vineyard (either on motorbike or walking) to reduce bird damage. Other methods included bird alarms, airhorns and sirens mounted on motorbikes, scarecrows, kites and balloons. See Table 3 below.

15 Table 3 Methods of Bird Control Identified in Personal Interviews

Method Description 1 Motorbike and Shotgun 2 Motorbike with Mounted Siren 3 Motorbike with Mounted Airhorn 4 Gas Gun 5 AV – Alarm 6 Multi-Frequency Generator 7 Netting 8 Traps 9 Balloons 10 Tape (i.e. strung along rows) 11 Scarecrows 12 Kites 13 Walking Patrols 14 Walking Patrols with Shotgun

Bird Damage Bird damage appeared to be the biggest problem facing growers in the 1998 season. The dry weather has been good for grape growers and levels of fungal diseases have been low. Some believed that there has been a higher population of birds in 1998, possibly due to the dry 1997 winter, although others believed the bird population has been lower because of the drought. The main birds mentioned as a problem were waxeyes and starlings. Waxeyes are self-introduced birds that are protected.

The problem occurs from approximately Waitangi day each year, when veraison8 occurs, until the end of April. Bird damage is a seasonal problem with grapes, but when considering other crops such as apples and cherries, the problem can be all year round.

When asked to estimate the level of bird damage that still occurred, answers ranged from less than 1% to 20%, although they were mainly less than 5%. The growers were also asked to estimate damage if they did not employ any bird control measures. These answers ranged from 10% to 100%. The majority of the answers were less than 50%.

8Veraison is when the sugar level in the grape begins to rise.

Birds tend to pick early maturing fruit, and will continue to eat that variety. The birds also prefer red grape varieties. Growers mentioned they had moved away from bulk variety grapes such as Muller Thurgau, to varietal grapes such as Chardonnay, which contain more sugar. Birds are more attracted to higher sugar levels so the level of damage is greater. Because these varietal types of grapes have a higher value, the cost of the damage is also greater. Hence, the secondary effects of bird damage are more significant.

Netting One grower used only netting on the vineyard but the vineyard still suffered considerable damage from birds. However, others growers did not use netting due to its high initial outlay. These growers would only consider netting as a ‘last resort’. Some growers mentioned that they would probably use more netting in the future, although a few thought that more research was needed into the negative aspects of netting. These include shading and the fact that netting interferes with spraying. It was also suggested that netting delays ripening by up to 10%, especially with red varieties. Grapes need high sunlight levels on their leaves to increase quality. Humidity can also be a problem. If the season has a high humidity level, or sunlight levels are low, the quality of grape can be reduced. Netting can intensify these effects.

Shelter Belts and Trees Areas of the vineyard with shelter belts and trees are the worst affected. Many shelterbelts and trees had been removed where possible. Shelterbelts are considered unnecessary, although they help ripen the fruit quicker, less bird damage seems to be preferred. One grower mentioned that they had left a shelterbelt to protect young vines from the wind.

Price Effect Some respondents identified that there is the potential for bird damage to have a price effect. In an acute situation, where damage is severe, grape buyers will either pay less, or not purchase the fruit from individual vineyards. From an industry perspective, the decreased supply due to damage could increase the price of undamaged fruit. Possibly, it is only when the majority of vineyards suffer severe damage that there will be a price effect across the industry. However, severe damage mainly occurs when there is a ‘bad season’, and fungal diseases are rife. 1998 was considered a ‘good’ year, so a price effect was not considered.

Alternative Methods In terms of alternative methods, some were mentioned that were not currently used by those interviewed. These included: Savannah, a natural bird deterrent that is a mixture of garlic and chilli etc, alphachloralase and Measurol. Measurol is now banned.

There were two alternative methods mentioned that the growers actually used. Hawks are considered by several growers to be an effective deterrent to birds. One grower had built perches and left food to encourage them to the area. Another planted Riesling,

17 which birds apparently dislike, around the edges of the vineyard, particularly where shelter belts and trees are serious problems. Others mentioned that human presence on the vineyard is an effective deterrent.

Some growers advocated an industry wide post harvesting strategy. Mechanical harvesting means that there are bunches of grapes left on the vines. This provides extra food for birds. Some thought that these bunches should be picked and dumped to reduce the food supply. An industry wide strategy would involve other crops such as sweetcorn, cherries, apples and olives. Such a strategy may help reduce the bird population. But even if the local population is reduced, some believe that birds will just come from further away.

Social Issues Bird control becomes a social issue because of the ‘noise’ pollution caused by some bird scaring devices such as gas guns and shotguns. Noise is a sensitive issue that is difficult and expensive to deal with. Growers have a right to protect their crops and neighbours have a right to ‘peace’. Isolated vineyards cause no problems, only those situated near urban habitation.

Most of the growers used gas guns and shotguns, although, as mentioned above, one used only netting. One respondent suggested that more shooting would be employed if gas guns were disallowed and another found shooting an inefficient use of time. It is a common story for birds to be seen perched on gas guns, which leads some to question their effectiveness. It has also been suggested that they cost more than they save growers in terms of crop damage. This is difficult to assess because growers commonly use a combination of methods, which is more effective. It seems that the majority of growers choose to use gas guns, this suggests that the method is cost effective. Further study into bird behaviour could be helpful when considering the direct effectiveness of gas guns.

Gas guns can be a significant problem, particularly if they are not operated properly and left on all night, for example. The Marlborough Grape Growers Association’s suggested limit on gas guns was quoted as 12 shots per hour by one of the interviewees. These voluntary guidelines have worked well in the past, but as the noise issue is of increasing magnitude, there is now a move towards having ‘rules’. Another believed that the Council had a time restriction on when gas guns could be used, i.e. between 6am and 6pm and on weekdays only. There were rules similar to this set out in the Proposed Wairua/Awatere District Plan. Tape can also be a noise ‘nuisance’ as it whistles when blowing in the wind. Under the Resource Management Act, landowners are obligated to take the best practical action to ensure noise levels are not unreasonable. If they don’t, they can be served with noise abatement notices. The difficulty is defining what ‘unreasonable’ is.

During the preparation of their Annual Plan, the Marlborough District Council received about 30 submissions on the issue of noise, ranging from one extreme to the other, all of which must be reviewed. At the time of these interviews the Council was in the

18 process of finalising their Annual Plan. The Council cannot ban outright bird scaring devices in rural areas, but can aim to have a workable rule that can be implemented and enforced. The difficulty in finding a workable rule is the measurement of noise (there are about 127 different methods) and understanding the problem of bird damage which is essential to achieving an effective management regime.

In order to understand the problem, the economics of bird control need to be considered. The question that needs to be addressed is “are the quantifiable benefits of ‘noisy’ bird control measures greater than the loss without them?” However, evaluation of policy measures must also address the issue of distribution of these costs and benefits. These equity consequences are not dealt with in economic analysis, which only concerns itself with the question of efficiency.

3.1.3 The Postal Survey Selection of Sample The sample population for the survey consisted of a combined list of members of the Marlborough Grape Growers Association and the Wine Institute of New Zealand. The sample size was 249, but it was later found that there were approximately 200 vineyards on the region. Basically, the sample covered the whole population of growers in the Marlborough Region. The difference in the numbers is probably due to ‘double ups’ through combining the two lists.

Development of the Questionnaire The survey needed to specify that the information given should relate to a defined vineyard area (given by the respondent) and the 1998 season. It was decided that in terms of general information, only vineyard size, crop value and crop weight were necessary in the postal survey.

The bird control methods identified in Table 3 were selected for the postal survey and included as a table for simplicity. The postal survey also included a further option of ‘other’ for any extra methods employed by the vineyard. See Appendix 1 for a copy of the actual survey. The participant was asked to tick each method used and also give an ‘effectiveness rating’ for that individual method. The rating system used in the personal interviews appeared to work effectively, so it was used again in the postal survey. However, the rating for frequency of use and overall ranking were removed because it was decided they would have been difficult to quantify into meaningful results.

The participants were then asked for information regarding the cost of each method employed. The questions were very specific in order to gather easily quantifiable and comparable information (as discussed above).

Finally, the questionnaire asked for an estimate of bird damage that occurs when the vineyard employs bird control and an estimate of bird damage if the vineyard had no

19 bird control. This was to determine the level of remaining damage and the perceived amount of damage avoided by bird control. Although it was identified that each vineyard has problem areas where bird damage is worse, it was felt that only one overall vineyard estimate of bird damage was necessary for each of these questions.

It was considered unnecessary to ask about the frequency and magnitude of bird damage across years in the postal survey because most of the answers from the personal interviews were the same. That is, bird damage occurs every year for a period of approximately 2 ½ months and varies in magnitude from year to year.

Unfortunately, respondents were not asked if their answers included Good and Services Tax (GST) or not. Due to this oversight, we have to assume that they did, unless otherwise stated. This will slightly understate costs in cases where they did not. The crop values are assumed to be GST exclusive9.

A summary of the research results was offered as a ‘reward’ to those who participated, so respondents were asked if they would like a copy of the results. This was the final question.

Survey Response Rate Response to the postal survey was quite high. In total, 249 postal surveys were mailed out, however many of the addresses were to the same vineyard, while some vineyards have had no crop over the 1998 season. Table 4 below categorises the responses that were received.

Table 4 Breakdown of Respondents

Completed Surveys Returned 125 Returned to Sender 4 Sold, leased or no vineyard 6 No harvest in 1998 8 Same as another return 15 Other 4 Total No of Respondents 162

9 Using figures from the BNZ Wine and Grape Industry Statistical Annual 1998, the average crop value per vineyard in the Marlborough region (assuming there are 200 vineyards) is $160,540 excl GST. In this analysis the average crop value per vineyard was $161,440. As these values are very close, it is assumed that the crop values given in the survey did not include GST.

20 The 1998 Vineyard Survey10, which is estimated to cover 95% of the national vineyard11, found a total of 189 vineyards in the Marlborough region. The 125 completed returns account for 66% of these vineyards, although, it should be considered that there are probably just over 200 vineyards in the region. Therefore the response rate is approximately 62%.

In terms of area coverage, the total (producing) vineyard area in Marlborough is estimated to be 2747 Ha12, while the total area covered by the responses is 2201 Ha. This accounts of approximately 80% of the area. The 1998 Vineyard Survey found an average vineyard size of 14.5 Ha13, while this survey found an average of 17.6 Ha. The medians were 7.8 Ha and 10.1 Ha respectively. This suggests that many of the non- respondents to this postal survey have smaller vineyards. Hence, their absence produces a higher average. This has some implications for the results that will be discussed in the section on Limitations.

3.2 Evaluation of Costs

The following assumptions were made in order to complete this analysis.

3.2.1 Methods of Evaluation Costs were evaluated based on the figures given by the respondents. That is, for each method of bird control, the capital cost, maintenance, labour and vehicle costs etc were added to attain the total cost. Where respondents entered a question mark, an average of the other respondents’ values was used. However, this was not done for the estimates of damage. Where a range was given, the values were averaged.

If a shotgun was used for both Method 1 (motorbike patrols) and Method 14 (walking patrols), the cost of the shotgun was divided evenly between the two methods.

The current Goods and Services tax (GST) rate of 12.5% was also assumed. This has been the GST rate since 1988. As the average years of purchase for capital equipment were all later than this date, this was the most appropriate rate to use.

3.2.2 Estimates of Hourly Running Costs and Labour Labour The hourly rate is based on the typical wage rate of Orchard Workers in Canterbury. These estimates, given in Table 5 below, were considered to be comparable to wages on a Vineyard in Marlborough. The Grape Growing ACC premium of $0.217 ($2.26 per $100) was added to the average of $9.59.

10 Conducted by Hort+Research 1998 11 The Bank of New Zealand Wine & Grape Industry Statistical Annual 1998, p3 12 ibid., p9 13 ibid., p10

Table 5 Wages and Salaries for Horticultural Employees

Orchard Workers in Canterbury Hourly Rate Permanent Staff Permanent Leading Hand $10.38 to $10.71 Permanent Leading Hand II $9.64 to $10.01 Casual Staff Casual $7.54 to $8.52 Casual Supervisor $10.00 Average of Above $9.59 These figures exclude Holiday Pay Adapted from the Financial Budget Manual 1998

Hence, an hourly rate of $9.85 was used to estimate the cost of labour. This rate was used for all of the labour hours, including hours worked by the vineyard manager and staff on salaries. The actual wage will vary depending on training, experience and alternative opportunities (Efferson, p70), however, it is assumed that workers receive $9.85 per hour, and that the manager and those on salaries would at least expect the same amount as if he/she worked for a neighbouring vineyard.

Where a contractor had provided the shotguns for Method 1 and 14, a contractor rate of $12.05 per hour was used. This figure is based on the hourly labour rate of $9.85, plus a shotgun allowance of $2.20 per hour.

An income tax rate of 21.5% was used, as this was the general tax rate for income between $0 and $34,200 per annum for the period of the analysis.

Tractor The hourly running cost of a tractor was based on figures given in the Financial Budget Manual 1998. An average of the estimates given in Table 6 was used.

22 Table 6 Total Tractor Running Costs

Tractor Type Total fixed cost per Total variable costs Total cost hour (580 hours/year) per hour excluding per hour labour 75kW (100HP) $18.46 $17.03 $35.49 2WD Tractor 75kW (100HP) $21.53 $18.68 $40.21 4WD Tractor Average $42.58

Motorbike The estimate of a motorbike running cost was based on a number of figures in the Financial Budget Manual. However, unlike the tractor estimate, they were not as explicit. The cost was estimated at $10 per hour. UtilityVehicle The cost of a Ute per hour was based on the assumption that the vehicle travelled at 25 km/hour, and travelled between 1601 and 2000 kilometres per year. The cost per kilometre is estimated at 77.9 cents (Financial Budget Manual, 1998, pB-112). Hence, the cost per hour is $19.47.

3.2.3 Evaluation of Capital Equipment To evaluate the cost of capital equipment for one year of use, an equivalent annual annuity was used. This was based on the assumption that the vineyard took out a table mortgage to pay for the equipment, the equivalent annuity being the value of one year’s payment.

The interest rates used for each year were an average of the 12 months in that year. For 1998, the interest rate was an average of the first 9 months of the year.

Chapter 4

Analysis of Results

4.1 Methods of Bird Control Used and their Effectiveness Ratings The methods of bird control used by the respondents and the effectiveness ratings given are discussed below. Note that the respondents were asked to rate each method individually, which may have been difficult in situations where many methods are used and combined. However, the ratings give an idea of the perceived effectiveness of each.

Method 1: Motorbike and Shotgun This method was used by 68% (85) of the respondents. The effectiveness ratings given by each are shown in Figure 1 below. Of these 85 users, 24 combine this method with Method 2 and/or Method 3; a motorbike with mounted siren or airhorn respectively. Figure 1 below shows how the respondents rated Method 1 in terms of effectiveness. The distribution of ratings is significantly skewed to the right, which indicates that this method is rated quite highly.

Figure 1 Frequency of Effectiveness Ratings for Method 1

0 1

Don’t Poor Below Average Above Ex c ellent know Average Average Effectiveness Rating

Method 2: Motorbike with Mounted Siren This method was only used by 5% (6) of the respondents. Due to this, and the similarity to Method 3, the two were combined. The results are discussed below.

Method 3: Motorbike with Mounted Airhorn This method was more popular than Method 2, with 28% (35) of the respondents employing this method. The combined results of Methods 2 and 3 are below.

Method 2 & 3: Motorbike with Mounted Siren or Airhorn There were a total of 40 respondents using either one of these methods. One of the respondents used both methods, so there are 41 observations (33%) on the following histogram (Figure 2) showing the associated effectiveness ratings. This distribution is also skewed to the right, so this method is also rated quite highly.

Figure 2 Frequency of Effectiveness Ratings for Method 2 & 3

Effectiveness Rating

1 0 1 Don’t know Poor Below Average Above Ex c ellent Average Average

Method 4: Gas Gun This method had the highest level of use with 77% (96) of the respondents using the method. Figure 3 below shows that the distribution of ratings is fairly evenly spread, although it is very slightly skewed to the left. This indicates that the method is perceived as being generally average in effectiveness.

25 Figure 3 Frequency of Effectiveness Ratings for Method 4

19 18

9 1 3

Don’t know Poor Below Average Above Ex c ellent Average Average

Effectiveness Rating

Method 5: AV-Alarm Of the 125 respondents, 34 (27%) of respondents used this method. It can be seen in Figure 4 that opinions on the effectiveness of AV-Alarms vary widely, and in general, the method did not rate highly.

Figure 4 Frequency of Effectiveness Ratings for Method 5

11 10

2 2 1